Archive ouverte HAL-LAAS
Not a member yet
12189 research outputs found
Sort by
A Nonlinear MPC for Physical Human-Aerial Robot Interaction in Collaborative Transportation Tasks
No full textInternational audienceAerial robots are transitioning from traditional surveillance and monitoring roles to more advanced tasks involving physical interaction. Despite this progress, physical Human-Aerial Robot Interaction remains largely underexplored due to the complexity and stability-related issues of such platforms. This paper introduces a novel control framework that enables an aerial platform to cooperatively transport an object with a human operator. The control approach is built on a nonlinear model predictive control (NMPC), integrating the dynamic models of the human, the aerial robot, and the transported object. To ensure safe and robust physical interaction, the NMPC is combined with a compliant controller. Additionally, our controller prioritizes forward motion over lateral movements to accommodate the human's natural direction of motion. We validate this framework through indoor flight experiments, demonstrating how a human operator and a fully actuated hexarotor can effectively collaborate to transport a bar. The results highlight the aerial robot's ability to assist the human during physical transportation tasks, enhancing efficiency and comfort
Openprobe: a frugal multiparameter probe for marine and continental waters monitoring
No full textInternational audienceAquatic ecosystems need accurate observation to understand its trajectory in a context of increasing anthropogenic pressure and climate change. These environments are however undersampled in many areas (Brewin et al. 2017, Kirchner et al. 2004): acquiring essential variables at relevant spatial and temporal resolution is often not compatible with traditional methods of sample collection followed by laboratory analysis, but field-deployable or in-situ systems are either too expensive, too complicated to be used by non-trained users or even non-existent. In an effort to propose an alternative in-situ measurement system for biogeochemical and physicochemical measurement in marine and continental waters, we developed a multiparameter probe that measures 7 parameters for a very modest bill-of-materials cost (around 300 € for a single prototype). The performance of this probe is sufficient for most water bodies, both in terms of range and sensitivity, and its replication is facilitated by the use of rapid-prototyping methods. The Openprobe multiparameter (Fig. 1) is able to measure conductivity, temperature, pressure (depth or water level above), dissolved oxygen, turbidity, chlorophyll a, photosynthetically active radiation (PAR) as well as water color. These parameters were chosen based on a compromise between their relevance and our capability to measure them at a scientifically appropriate standard using low-cost components, an approach which resonates with the definition of the Standard Observations in eLTER (Zacharias et al. 2021). The electronic architecture is based around an Adafruit Feather M0 microcontroller board, an Adafruit Adalogger microSD-card logger with a Real-Time-Clock, and two custom Printed Circuit Boards (PCBs) that hosts the different sensors. Housing of the probe uses a Blue Robotics 2" diameter enclosure tubes. The sensor's PCBs are integrated in end-caps fabricated by stereolithography (SLA), using a desktop Formlabs Form 3 SLA printer with Black V4 resin. Compared to conventional fused-deposition modelling 3D printing (FDM), SLA is an isotropic printing technique which delivers fully dense and watertight parts. The higher resolution of SLA also allows to create o-ring grooves with appropriate tolerance, and has shown to be a valid fabrication technique even for the challenging pressures encountered in deep-sea science (Karp et al. 2023, Phillips et al. 2019). Atop of that, the 3D printed end-cads ensures auto-alignment of the optical elements (LEDs, photodiodes) and optical filters. The assembly/disassembly can be done manually without any tool, for easy access to the electronics, the battery, or the micro SD card on the field. Polydimethylsiloxane (PDMS), a silicone-based material with good transparency in the VIS-IR, is used within the end-caps to create optical ports, and also acts similarly to a potting material to ensure watertightness. The first PCB is integrated at the top of the probe, and contains the CTD (Conductivity, Depth, Temperature) as well as the PAR sensor. The conductivity sensor is based around a two-electrode cell on a ceramic substrate connected to an impedance analyzer integrated circuit (IC). Intercomparison with a Decagon CTD-10 and an Atlas Scientific EZO sensor showed that our sensor offered a better sensitivity, together with a smaller cost and a lower power consumption. Pressure and temperature sensors are based around the same IC used by Blue Robotics pressure and temperature sensors, and as such as been validated in many conditions (Poulsen et al. 2022). We worked on their integration on a custom PCB, with special care taken in minimizing the thermal inertia around the temperature sensor to reduce its response time when performing profiling analysis. The PAR sensor completes this part of the probe. It is based around the AMS AS7341, modified with a diffuser cut precisely with the xurography technique (Bartholomeusz et al. 2005), and intercompares with a+/- 5% confidence interval versus an Apogee SQ-512-SS sensor (Fig. 2), a value commonly exceeded when comparing different commercial sensors (Long et al. 2012). The second PCB, located at the opposite end of the probe, contains a chlorophyll a fluorometer, an oxygen optode, and a turbidity sensor. The chlorophyll a fluorometer uses synchronous detection: this approach allows efficient ambient light rejection, and captures the weak fluorescence signals observed due to the small fluorescence yield of in-vivo chl.a (Morrison 2003). The oxygen otpode uses PreSens Pst3 oxygen sensor spots, paired with a custom optoelectronic readout architecture. A DDS (Direct Digital Synthesizer) generates a sine wave to excite the Pst3 luminophore, and an I/Q demodulator measures the phase shift between excitation and the emitted fluorescence. The fluorescence signal is quenched in the presence of oxygen, which translates to a decrease of the fluorescence decay and in-fine by a phase shift . This is similar to the operating principle of Aanderaa optodes, commonly found in the Biogeochemical Argo floats. Our optode is intercompared with a benchtop PreSens Fibox 4 system and shown good accuracy (Fig. 3). Finally, the turbidity sensor implements 90° nephelometry as well as 180° backscattering measurements to cover a broad range of turbidity. Its architecture is based around an Analog Devices ADPD1080, an IC usually found in smartwatches for the measurement of photoplethysmography (PPG). The Openprobe multiparameter probe could be a very valuable tool in order to increase spatial and temporal resolution in aquatic ecosystem monitoring. Through the use of advanced, yet affordable electronic components initially developed for the Internet Of Things market, we created a frugal tool without compromising on accuracy
Infinite-Horizon Value Function Approximation for Model Predictive Control
No full textModel Predictive Control has emerged as a popular tool for robots to generate complex motions. However, the real-time requirement has limited the use of hard constraints and large preview horizons, which are necessary to ensure safety and stability. In practice, practitioners have to carefully design cost functions that can imitate an infinite horizon formulation, which is tedious and often results in local minima. In this work, we study how to approximate the infinite horizon value function of constrained optimal control problems with neural networks using value iteration and trajectory optimization. Furthermore, we demonstrate how using this value function approximation as a terminal cost provides global stability to the model predictive controller. The approach is validated on two toy problems and a real-world scenario with online obstacle avoidance on an industrial manipulator where the value function is conditioned to the goal and obstacle
Event triggered control and exponential stability for infinite dimensional linear systems
No full textInternational audienceThis article aims at providing a unified analysis of the exponential stabilization of some abstract infinite dimensional systems undergoing an event-triggering mechanism that samples the control input. The partial differential equation is supposed to be defined by a skew-adjoint operator and controlled and observed through bounded operators. The continuously controlled closed loop system is assumed to be exponentially stable and the goal is to prove that a well-designed event-triggering mechanism to rule the time updates of the sampled control will allow to keep such a stability property. The key of the proof relies on the existence of an adequate Lyapunov functional. Existence and regularity of the solution to the closed-loop event-triggered system are also proven, along with the avoidance of Zeno behavior
Ubiquitous IoT Through Space Communications
No full textNational audienceLow-Power Wide-Area Networks (LPWANs) provide a low-cost solution for connecting low-power devices over long distances. Among LPWAN technologies, Narrowband IoT (NB-IoT) has gained prominence because of its wide coverage and high reliability. Recently, as the demand for IoT connectivity in remote and underserved regions continues to grow, integrating NB-IoT with Low-Earth Orbit (LEO) satellite networks has gained increasing attention. Due to satellite communication's dynamic nature, this integration poses significant challenges, including synchronization issues and managing high-density User Equipment (UE) during the random access procedure. This digest paper summarizes key contributions from a doctoral thesis that addresses these challenges. A systematic framework based on key performance indicators (KPIs) is proposed to evaluate and optimize satellite IoT communication, addressing reliability, latency, throughput, and energy efficiency trade-offs. A lightweight downlink synchronization method is proposed, which reduces device complexity in LEO satellite environments. In addition, a GNSS-free wake-up strategy is developed to improve energy efficiency, particularly in scenarios with intermittent satellite coverage. To address the challenge of random access in high-density environments, an early detection method is introduced. This method reduces collisions during the random access procedure and improves the network capacity to support many UEs. These contributions provide a comprehensive set of solutions to address the limitations of current NB-IoT and LEO satellite integration strategies, paving the way for efficient and scalable IoT networks in challenging environments
Diodes lasers stabilisées par CRIGF accordables
No full textNational audienceLes filtres CRIGFs permettent de réaliser des systèmes de diodes laser en cavité externe stabilisés en fréquence compacts et stables. En intégration dans ces cavités des CRIGFs accordables réalisés en filière Niobate de Lithium, nous montrons qu’elles peuvent être accordées spectralement en exploitant les effets électro-optiques et thermo-optiques de ce matériau
ACPAC atelier du 19 mars 2025
No full textNational audienceTravaux des experts techniques et scientifiques de l'Action concertée Packaging du PEPR électronique lors des journées scientifiques annuelles 202
LinA: A faster approach to piecewise linear approximations using corridors and its application to mixed-integer optimization
No full textInternational audienceIn this paper, we address the problem of approximating and over/under-estimating univariate functions with piecewise linear (PWL) functions with the minimum number of linear segments given a bound on the pointwise approximation error allowed. Through a new geometric approach and building on the work of Ngueveu [Ngu19], we develop new algorithms that can solve the problem in quasi-logarithmic time on a very broad class of error types. Such algorithms find many applications, mostly related to solving certain classes of (mixed-integer) nonlinear and nonconvex programming (MINLP) problems by mixed-integer linear programming (MILP) techniques. An efficient implementation of our algorithms is available as a Julia package. Benchmarks are also provided to showcase how our method outperforms the state-of-the-art for this problem. Finally, we show how our algorithms can be used to efficiently solve certain classes of MINLP problems by a case study on multicommodity network design problems with congestion
Reinforcement Learning from Wild Animal Videos
No full textWe propose to learn legged robot locomotion skills by watching thousands of wild animal videos from the internet, such as those featured in nature documentaries. Indeed, such videos offer a rich and diverse collection of plausible motion examples, which could inform how robots should move. To achieve this, we introduce Reinforcement Learning from Wild Animal Videos (RLWAV), a method to ground these motions into physical robots. We first train a video classifier on a large-scale animal video dataset to recognize actions from RGB clips of animals in their natural habitats. We then train a multi-skill policy to control a robot in a physics simulator, using the classification score of a third-person camera capturing videos of the robot's movements as a reward for reinforcement learning. Finally, we directly transfer the learned policy to a real quadruped Solo. Remarkably, despite the extreme gap in both domain and embodiment between animals in the wild and robots, our approach enables the policy to learn diverse skills such as walking, jumping, and keeping still, without relying on reference trajectories nor skill-specific rewards
Alimentation autonome sans fil et sans batterie par électrothermogénération
No full textL'électrothermogénération est bien adaptée pour récupérer de l'énergie électrique en présence d'un gradient thermique permanent mais il y a toujours quelques problèmes techniques comme présenté dans cet article basé sur la réalisation d'une alimentation électrique destiné à un capteur d'O2 en milieu industriel.International audienc